Silicon dioxide@graphene oxide-graft-poly(γ-benzyl-L-glutamate) as an advanced hybrid nanofiller reinforces poly(L-lactide)

Abstract

A critical challenge in the preparation of reinforced polymer nanocomposites is to prevent the aggregation of nanofillers. In this work, a novel poly(γ-benzyl-L-glutamate) (PBLG)-modified silicon dioxide@graphene oxide nanofiller (SiO₂@GO-g-PBLG) was prepared via a continuous electrostatic complex and ring-opening polymerization (ROP) method. The grafted PBLG prevented the aggregation of SiO₂ nanoparticles and GO sheets, realized the colloid stability of SiO₂@GO-g-PBLG in organic solvents, and thus increased its phase interaction in a polymer matrix. The hybrid nanofiller greatly enhanced the mechanical properties of poly(L-lactide) (PLLA). As a typical feature, the tensile strength of PLLA nanocomposites with only 5 wt% of SiO₂@GO-g-PBLG was 88.9 MPa, about 45% higher than that of pure PLLA. In addition, the hybrid nanofillers also have some positive effects on improving the thermal stability of PLLA. Therefore, SiO₂@GO-g-PBLG was a promising hybrid nanofiller to reinforce polymers such as PLLA.

---

1. Introduction

The mixture of nanofillers with polymers to fabricate polymer nanocomposites has been widely investigated. Nanofillers play an important role in optimizing the mechanical, thermal, and electrical properties of polymer nanocomposites, and thus expand their application scopes. In the past decades, a variety of nanofillers, such as graphene, carbon nanotubes, silica nanoparticles, and glass fibers have been used to reinforce polymer matrices.^1,2 Among them, graphene is a promising nanofiller to enhance the electrical and thermal conductivities, and mechanical strength of a polymer matrix.^1,3,4 Various graphene nanocomposites have been prepared, while the graphene sheets aggregate easily in a polymer matrix due to the strong π–π interactions among them. Thus, many methods, including covalent and non-covalent surface modifications, etc., have been proposed to tune the surface properties of graphene.^5–7

One common approach to hinder the aggregation of graphene sheets is to prepare the inorganic nanoparticle–graphene hybrids.^8,9 When the inorganic nanoparticles are anchored on the surface of graphene, it can reduce the π–π interaction; on the other hand, the aggregation of inorganic nanoparticles can also be prevented. Up to now, many inorganic nanoparticle–graphene hybrids have been prepared.⁹ More interestingly, much better synergistic reinforcing effect toward polymer nanocomposites may appear and realize, when different types of nanofillers are simultaneously added into polymer matrix. Many hybrid nanofillers consisting of two or more components, such as halloysite–graphene,¹⁰ carbon nanotube–graphene,¹¹ and carbon nanotube–halloysite,¹² have been prepared and used to reinforce various polymer matrix, which have shown more satisfying results than their single component. However, these hybrid nanofillers are all inorganic components, there still lack a bridge to connect the inorganic nanofillers and the polymer matrix.

As mentioned above, owing to the poor dispersion or non-regular arrangement of nanofillers, the polymer nanocomposites always have disappointing mechanical properties and fall far below their theoretical values.^13,14 In order to further improve the mechanical performances of polymer nanocomposites, an effective method is to graft polymer chains on the surface of nanofillers and thus enhance the phase compatibility between nanofillers and polymer matrix.

Poly(L-lactide) (PLLA) is an important biodegradable polymer, which has been widely used in various biomedical fields or as a disposable package. However, the low mechanical properties have limited its final application; so many PLLA nanocomposites have been designed and prepared.¹⁵ For example, Chen's group prepared the PLLA-modified hydroxyapatite (HA-g-PLLA) and found that the tensile strength of HA-g-PLLA/PLLA nanocomposites can be improved to 75 MPa, which indicated much potential applications in bone tissue engineering.^16,17 Recently, the graphene/PLLA nanocomposites have also been reported, and some researchers also use the polymer-modified graphene to blend with PLLA.^18,19 For example, PLLA-grafted graphene oxide (GO-g-PLLA) was used to reinforce the PLLA matrix, and the tensile strength of GO-g-PLLA/PLLA could be about 75 MPa.²⁰ Up to now, it is still a challenge to acquire PLLA nanocomposites with sufficient mechanical strength.

In order to hinder the aggregation of nanofillers, enhance their phase compatibility with polymer matrix, and also obtain high mechanical strength of PLLA nanocomposites, a novel ternary hybrid nanofiller, i.e., SiO₂@GO-g-PBLG, was prepared. Typically, as shown in Fig. 1, silicon dioxide@graphene oxide (SiO₂@GO) was firstly prepared. And then, the biodegradable poly(γ-benzyl-L-glutamate) (PBLG) was grafted onto the surface of SiO₂@GO, forming a ternary organic–inorganic hybrid, that is, SiO₂@GO-g-PBLG. The newly prepared hybrid can prevent the aggregation of SiO₂ nanoparticle and graphene sheet, and realize the homogeneous dispersion in polymer matrix. More importantly, the PBLG chains can bridge SiO₂@GO and PLLA matrix, and thus the PLLA nanocomposites with excellent strength were obtained.

Fig. 1 Synthesis route of SiO₂@GO-g-PBLG.

2. Experimental section

2.1. Materials

GO was prepared by the oxidation of graphite according to the modified Hummers method.^21,22 Briefly, 2.5 g of natural graphite, 2.5 g of sodium nitrate (NaNO₃), and 7.5 g of potassium permanganate (KMnO₄) were mixed in a bottle, and then 50.0 mL of concentrated sulfuric acid (H₂SO₄) was slowly added under stirring at 5 °C. 1 h later, the mixture was heated to 35 °C, 100.0 mL of water was added into the mixture, and the temperature increased to about 90 °C. The reaction was maintained at 95 °C for 15 min. The mixture was poured into 300.0 mL of deionized water, after that 20.0 mL of hydrogen peroxide (H₂O₂) was added into the suspension. Finally, the solid product was collected by filtration and washed with 5% hydrogen chloride (HCl) aqueous several times.

γ-Benzyl-L-glutamate N-carboxyanhydride (BLG NCA) was synthesized according to the previously published method.²³ Tetraethylorthosilicate (TEOS) was purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, P. R. China). 3-Aminopropyltriethoxysilane (APTES) was obtained from Sigma-Aldrich (Shanghai, P. R. China). 1,6-Diaminohexane was purchased from Tokyo Chemical Industry Co., Ltd. (Shanghai, P. R. China). PLLA was donated by Zhejiang Hisun Biomaterials Co., Ltd. (Taizhou, P. R. China). All the chemicals were used without further purification.

2.2. Preparation of SiO₂@GO

Firstly, the amino-functionalized silica nanosphere was prepared according to the Stöber method.²⁴ Briefly, ammonia (9.0 mL), ethanol (60.0 mL), and deionized water (25.0 mL) were mixed together, and then TEOS (4.5 mL) was added into the mixture and stirred for 5 h. Subsequently, APTES (200 mg) was added to the reaction system and stirred for 24 h. The final product (SiO₂–NH₂) was collected by centrifugation, washed with ethanol, and dried under vacuum.

Secondly, SiO₂@GO was prepared by a simple electrostatic complex technique. Briefly, SiO₂–NH₂ (300.0 mg) and GO (700.0 mg) were dispersed in deionized water (300.0 mL) by ultrasonic and stirred at room temperature for 24 h. And then, the resultant product was obtained by centrifugation and washed with deionized water several times to remove the residual GO.

2.3. Preparation of SiO₂@GO-g-PBLG ternary hybrid

The novel SiO₂@GO-g-PBLG hybrid was prepared via the ring-opening polymerization (ROP) of BLG NCA. Firstly, SiO₂@GO and 1,6-diaminohexane was dispersed in water and stirred at 85 °C for 24 h. During this process, 1,6-diaminohexane reacted with the epoxy group on graphene sheet, and thus the amino-functionalized SiO₂@GO (i.e., SiO₂@GO-NH₂) was obtained. The final product was collected by centrifugation, washed with water, and then lyophilized.

To synthesize SiO₂@GO-g-PBLG, SiO₂@GO-NH₂ (250.0 mg) and BLG NCA (500.0 mg) were added in a dry flask under the protection of nitrogen (N₂), and then anhydrous N,N-dimethylformamide (DMF) (30.0 mL) was injected into the flask by syringe. The mixture was dispersed with ultrasound for 30 min, and then stirred at 35 °C for 48 h. The final product, that is, SiO₂@GO-g-PBLG, was separated by centrifugation, washed with DMF three times, and dried under vacuum.

2.4. Preparation of SiO₂@GO-g-PBLG/PLLA nanocomposites

To prepare the SiO₂@GO-g-PBLG/PLLA nanocomposites, a desired amount of SiO₂@GO-g-PBLG was dispersed in chloroform by ultrasonic treatment, and then PLLA sample was dissolved in SiO₂@GO-PBLG suspension. The nanocomposite solution was poured into a polytetrafluoroethylene mould to form a membrane by evaporation. The obtained membrane was dried under vacuum at 40 °C for 48 h, and annealed at 100 °C for 1 h. The obtained SiO₂@GO-g-PBLG/PLLA nanocomposites with different contents of nanofiller were prepared. The weight fractions of nanofillers were adjusted to be 0.5, 1, 2, and 5 wt%, and the corresponding nanocomposites were denoted as PLLA0.5, PLLA1, PLLA2, and PLLA5, respectively.

2.5. Characterizations

Morphologies and microstructures of the SiO₂@GO-g-PBLG/PLLA nanocomposites were investigated by scanning electron microscopy (SEM; FEI, QuanTA-200F, Eindhoven, The Netherlands) and transmission electron microscopy (TEM; JEOL, JEM-2100F, Japan), respectively. Fourier-transform infrared (FT-IR) spectra were recorded on a Fourier-transform infrared spectrophotometer (Shimadzu IR Prestige-21, Nakagyo-ku, Japan) using potassium bromide (KBr) disks. Thermal gravimetric analyses (TGA) was carried out under N₂ atmosphere from room temperature to 800 °C at a heating rate of 10 °C min⁻¹ with a thermogravimetric analyzer (Netzsch STA409PC, Selb, Germany). Tensile tests were carried out on an electromechanical universal testing machine (CMT, Shenzhen SANS Testing Machine Co., Ltd., Shenzhen, P. R. China) at a testing speed of 10 mm min⁻¹. Differential scanning calorimetry (DSC) was performed using a Shimadzu DSC-60 (Columbia, MD, USA) by a heat–cool–heat cycle between 0 and 200 °C with a heating and cooling rate of 10 °C min⁻¹.

3. Results and discussion

3.1. Preparation and characterization of SiO₂@GO-g-PBLG

As illustrated in Fig. 1, ascribed to the electrostatic attraction between SiO₂–NH₂ nanoparticle and GO sheet, the GO sheet and SiO₂–NH₂ nanoparticle spontaneously assembled into a new hybrid, as soon as they were mixed together. After the reaction with 1,6-diaminohexane, the amino group was attached onto the surface of SiO₂@GO to synthesize SiO₂@GO-NH₂, which could initiate the ROP of BLG NCA to form the ternary SiO₂@GO-g-PBLG hybrid. Compared with SiO₂@GO-NH₂, the presence of PBLG in SiO₂@GO-g-PBLG could be easily monitored by FT-IR through the newly appeared peaks at 1633 and 1550 cm⁻¹ (Fig. 2), which were characteristically assigned to the amide I and amide II in the PBLG backbone.

Fig. 2 FTIR spectra of GO, SiO₂, SiO₂@GO-NH₂, and SiO₂@GO-g-PBLG.

The TGA curves of GO, SiO₂, SiO₂@GO-NH₂, and SiO₂@GO-g-PBLG were shown in Fig. 3. The weight losses of different samples under 150 °C should be attributed to the volatilization of adsorbed water. As for GO and SiO₂, when the samples were heated to above 150 °C, the surface functional groups would decompose, and thus there was an evident weight loss for both GO and SiO₂@GO. Comparing the TGA curve of SiO₂@GO-NH₂ with the TGA of SiO₂@GO-g-PBLG, the effective weight losses of SiO₂@GO-NH₂ and SiO₂@GO-g-PBLG in the temperature range of 150 to 800 °C were 15.70 and 25.9 wt%, respectively, and these different values were derived from the decomposition of PBLG chain.

Fig. 3 TGA curves of GO, SiO₂, SiO₂@GO-NH₂, and SiO₂@GO-g-PBLG.

The morphologies of SiO₂@GO and SiO₂@GO-g-PBLG were observed by both SEM and TEM. As shown in the SEM micrographs of Fig. 4, the spherical structure of SiO₂ in SiO₂@GO was clear, and GO sheet was observed on the structure of SiO₂@GO nanoparticle. In addition, the size of GO sheet was much bigger than that of SiO₂, so it is possible that GO sheet could connect several SiO₂ nanoparticles together. After the polymerization of BLG NCA, PBLG chains were grafted on the surface of SiO₂@GO, resulting in the formation of polymer layer on the surface of SiO₂@GO. The TEM images of SiO₂@GO and SiO₂@GO-g-PBLG were also represented in Fig. 2. It can well demonstrated that GO was coated on the surface of SiO₂, meanwhile it also showed that GO sheet could connect SiO₂ nanoparticles together. As for the structure of SiO₂@GO-g-PBLG, the hydrophobic PBLG chains existed on the surface of SiO₂@GO, and thus altered the surface properties of SiO₂@GO. More importantly, the PBLG chains might be of vital importance when the hybrid was used to reinforce polymers, because this might increase the interactive area and phase compatibility between nanofiller and polymer matrix.

Fig. 4 SEM and TEM micrographs of SiO₂@GO and SiO₂@GO-g-PBLG.

As shown in Fig. 5, SiO₂@GO and SiO₂@GO-g-PBLG were added into a two phase solvents, i.e., the upper and lower layers were water and chloroform, respectively. After ultrasonic dispersion, SiO₂@GO dispersed homogeneously in the water phase, while SiO₂@GO-g-PBLG existed in chloroform and maintained its colloid stability for seven days. This result well demonstrated that the surface wettability of SiO₂@GO-g-PBLG was hydrophobic, which was much different with that of SiO₂@GO. So it may be very useful when SiO₂@GO-g-PBLG was blended with organic polymers. The PBLG chains might enhance the phase interaction between the nanofiller and polymer matrix, and thus obtain better mechanical properties.

Fig. 5 Photographs of SiO₂@GO and SiO₂@GO-g-PBLG in water/chloroform mixture.

3.2. Mechanical and thermal properties of nanofiller/PLLA nanocomposites

PLLA is a promising biodegradable polymer in the field of biomedical and civilian areas. It is of vital importance to obtain the PLLA nanocomposites with better mechanical properties. Fig. 6 shows the representative stress–strain curves of PLLA and PLLA nanocomposites with different SiO₂@GO-g-PBLG contents, and the corresponding tensile strength, tensile modulus, and elongation at break were shown in Table 1. The tensile strength and modulus of pure PLLA were 64 ± 3.0 MPa and 2.4 ± 0.3 GPa, respectively. The blend of hybrid nanofillers, i.e., SiO₂@GO-g-PBLG, greatly improved the strength of PLLA. As shown in Table 1, even 0.5 wt% of SiO₂@GO-g-PBLG could realize a great reinforcing effect on PLLA. The tensile strength of PLLA0.5 was 75.2 ± 2.8 MPa and about 17.5% higher than that of PLLA. The tensile modulus of PLLA0.5 was 2.9 ± 0.3 GPa and about 21% higher than that of pure PLLA. With the increase of SiO₂@GO-g-PBLG content, both the tensile strength and tensile modulus of PLLA nanocomposites were improved a lot. Especially, when the nanofiller content was higher up to 5 wt%, the PLLA5 nanocomposite showed excellent tensile strength at about 88.8 ± 3.2 MPa, at the meantime, the tensile modulus increased to 3.5 ± 0.3 GPa, which was about 45% higher than that of PLLA. All these results demonstrated that the novel hybrid SiO₂@GO-g-PBLG could work as an effective nanofiller to reinforce PLLA matrix. As a control, when the PLLA nanocomposite with 5 wt% of SiO₂@GO was prepared, its tensile strength was only 70.5 ± 4.3 MPa and much lower than that of SiO₂@GO-g-PBLG-filled PLLA nanocomposite. The results showed that the PBLG chain might greatly enhance the phase compatibility and interaction between the nanofiller and polymer matrix, and thus increase the mechanical properties of PLLA. In addition, when compared with the published results, the synergistic effect of SiO₂ and GO was more positive. As reported by Yin's group, the tensile strength of PLLA composites filled with 5 wt% PLLA-grated SiO₂ was about 67 MPa, while the composites filled with 5 wt% SiO₂ was only 56 MPa.²⁵ Xu's group prepared GO-g-PLLA, and the tensile strength of GO-g-PLLA/PLLA and GO/PLLA composites were about 72 and 53 MPa, respectively.²⁰

Fig. 6 Tensile stress–strain curves of PLLA and SiO₂@GO-g-PBLG/PLLA nanocomposites with different SiO₂@GO-g-PBLG contents.

Table 1 Mechanical properties of PLLA and its nanocomposites with different SiO₂@GO-g-PBLG contents

Samples	Tensile strength (MPa)	Modulus (GPa)	Elongation at break (%)
a PLLA-SG5 is a control composite sample with 5 wt% of SiO2@GO.
PLLA	64.0 ± 3.0	2.4 ± 0.3	4.3 ± 0.6
PLLA0.5	75.2 ± 2.8	2.9 ± 0.3	3.5 ± 0.5
PLLA1	79.0 ± 1.8	3.2 ± 0.10	3.4 ± 0.1
PLLA2	83.7 ± 4.5	3.3 ± 0.2	3.4 ± 0.2
PLLA5	88.8 ± 3.2	3.5 ± 0.3	3.6 ± 0.2
PLLA-SG5^a	70.5 ± 4.3	3.1 ± 0.4	2.8 ± 0.1

---

The morphologies of the tensile fracture surfaces for both PLLA and different PLLA nanocomposites were shown in Fig. 7. The fracture surface of PLLA was smooth, while the surfaces for PLLA nanocomposites were a little coarser than that of PLLA. Furthermore, some fibrous structure appeared on the fracture surfaces of PLLA nanocomposites, which might result from the reinforcement of SiO₂@GO-g-PBLG. To more clearly observe the surface morphology, a high magnification microimage of the fracture surface of PLLA5 was shown in Fig. 7. It showed clearly that the hybrid nanofillers dispersed homogeneously in the polymer matrix without the aggregation of nanoparticle. The phenomenon might be because that the phase compatibility between the SiO₂@GO-g-PBLG nanofiller and PLLA matrix was much better. Furthermore, as denoted by the arrows in Fig. 7, the interface between nanoparticle and polymer was obscure, and some wrinkles were found between the nanoparticle and polymer. The results implied that the phase interaction was much stronger, and this point might be an explanation for the improved mechanical properties as shown in Table 1.

Fig. 7 Fracture surface morphologies of PLLA and its nanocomposites with different SiO₂@GO-g-PBLG contents.

The thermal properties of SiO₂@GO-g-PBLG/PLLA nanocomposites were also investigated by TGA and DSC. As shown in Fig. 8 and S1, ESI,† lower SiO₂@GO-g-PBLG contents (e.g., 0.5 and 1 wt%) had little effect on the thermal stability of PLLA. When the content of nanofiller was more than 2 wt%, the thermal stability of PLLA nanocomposite was better than that of pure PLLA. On the other hand, the hybrid nanofiller showed quite slight effect on the melting temperature (T_m) and crystallization temperature (T_c). As shown in Fig. 9, the T_m of PLLA was 177.8 °C, while those of the PLLA nanocomposites were all in the range of 177.2–177.7 °C. The T_c of PLLA and its nanocomposites exhibited no significant difference. All the T_c values were in the range of 106.6–107.5 °C, and the T_c of PLLA5 sample was only 0.5 °C higher than that of PLLA. Furthermore, the SiO₂@GO-g-PBLG nanofillers might greatly enhance the crystallization speed of PLLA, when isothermally crystallized in a higher temperature. These results will be discussed in our future work.

Fig. 8 TGA decomposition graphs of pure PLLA and its nanocomposites with different SiO₂@GO-g-PBLG contents.

Fig. 9 DSC thermal graphs of PLLA and its nanocomposites with different SiO₂@GO-g-PBLG contents during the heating (A) and cooling process (B).

4. Conclusion

This work introduced an efficient method to prepare a ternary organic–inorganic hybrid, i.e., SiO₂@GO-g-PBLG. The hybrid can be dispersed in chloroform, and its phase compatibility with hydrophobic polymers should be better than those of pure inorganic nanofillers. In addition, the novel SiO₂@GO-g-PBLG exhibited the synergistic effect of different nanofillers and realized the homogenous dispersion of SiO₂ nanoparticle and GO sheet in PLLA. As a result, SiO₂@GO-g-PBLG greatly enhanced the mechanical properties of PLLA nanocomposites, implying its potential application in polymer nanocomposites.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (No. 51203073 and 51463013) and the Natural Science Foundation of Jiangxi Province of China (No. 20142BAB203018 and 20151BAB206011).

Notes and references

1. K. Hu, D. D. Kulkarni, I. Choi and V. V. Tsukruk, Prog. Polym. Sci., 2014, 39, 1934–1972.
2. X. Sun, H. Sun, H. Li and H. Peng, Adv. Mater., 2013, 25, 5153–5176.
3. X.-Z. Tong, F. Song, M.-Q. Li, X.-L. Wang, I.-J. Chin and Y.-Z. Wang, Compos. Sci. Technol., 2013, 88, 33–38.
4. Y. T. Park, Y. Qian, C. Chan, T. Suh, M. G. Nejhad, C. W. Macosko and A. Stein, Adv. Funct. Mater., 2015, 25, 575–585.
5. T. Kuilla, S. Bhadra, D. Yao, N. H. Kim, S. Bose and J. H. Lee, Prog. Polym. Sci., 2010, 35, 1350–1375.
6. T. Kuila, S. Bose, A. K. Mishra, P. Khanra, N. H. Kim and J. H. Lee, Prog. Mater. Sci., 2012, 57, 1061–1105.
7. P. Song, Z. Cao, Y. Cai, L. Zhao, Z. Fang and S. Fu, Polymer, 2011, 52, 4001–4010.
8. C. Zhang and T. Liu, Chin. Sci. Bull., 2012, 57, 3010–3021.
9. L. Na, C. Minhua and H. Changwen, Nanoscale, 2012, 4, 6205–6218.
10. Z. Tang, Q. Wei, T. Lin, B. Guo and D. Jia, RSC Adv., 2013, 3, 17057–17064.
11. C. Zhang, S. Huang, W. W. Tjiu, W. Fan and T. Liu, J. Mater. Chem., 2012, 22, 2427–2434.
12. L. Jiang, C. Zhang, M. Liu, Z. Yang, W. W. Tjiu and T. Liu, Compos. Sci. Technol., 2014, 91, 98–103.
13. Y. Dzenis, Science, 2008, 319, 419–420.
14. P. Podsiadlo, A. K. Kaushik, E. M. Arruda, A. M. Waas, B. S. Shim, J. Xu, H. Nandivada, B. G. Pumplin, J. Lahann, A. Ramamoorthy and N. A. Kotov, Science, 2007, 318, 80–83.
15. J.-M. Raquez, Y. Habibi, M. Murariu and P. Dubois, Prog. Polym. Sci., 2013, 38, 1504–1542.
16. Z. K. Hong, P. B. Zhang, C. L. He, X. Y. Qiu, A. X. Liu, L. Chen, X. S. Chen and X. B. Jing, Biomaterials, 2005, 26, 6296–6304.
17. P. Zhang, Z. Hong, T. Yu, X. Chen and X. Jing, Biomaterials, 2009, 30, 58–70.
18. Z. Qiu and W. Guan, RSC Adv., 2014, 4, 9463–9470.
19. P. Chen, Y. Wang, T. Wei, Z. Meng, X. Jia and K. Xi, J. Mater. Chem. A, 2013, 1, 9028–9032.
20. W. Li, Z. Xu, L. Chen, M. Shan, X. Tian, C. Yang, H. Lv and X. Qian, Chem. Eng. J., 2014, 237, 291–299.
21. W. S. Hummers and R. E. Offeman, J. Am. Chem. Soc., 1958, 80, 1339.
22. D. Chen, X. Wang, T. Liu, X. Wang and J. Li, ACS Appl. Mater. Interfaces, 2010, 2, 2005–2011.
23. Y. Yao, W. Li, S. Wang, D. Yan and X. Chen, Macromol. Rapid Commun., 2006, 27, 2019–2025.
24. W. Stöber, A. Fink and E. Bohn, J. Colloid Interface Sci., 1968, 26, 62–69.
25. S. Yan, J. Yin, Y. Yang, Z. Dai, J. Ma and X. Chen, Polymer, 2007, 48, 1688–1694.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra27104e

‡ These authors contributed equally to this work.

---